3D Reconstruction of Underwater Objects Considering Refraction: Challenges and Solutions

Yuri Rzhanov and Igor Kozlov


Friday, Mar. 23, 2018, 3:10pm
Chase 130

3D reconstruction of scenes from imagery is a well-researched topic. However, this is applicable only to images taken in air. Underwater imagery poses new challenges due to the fact that most cameras are designed to operate in air, and in water they have to be placed in protective housings. Housing vieweport distorts imagery by refracting light at two surfaces separating media with different refraction indexes. Correction of imagery for Euclidean 3D reconstruction required additional calibration step - determination of "refractive" parameters. These parameters depend on the geometry of a viewport. Most widely used viewports are flat and hemispherical. Refractive parameters are very difficult to measure directly and are known to be notoriously difficult to be determined by computational methods. We propose novel approaches for determination of these parameters and show how 3D reconstruction with refraction taken into account improves the results.


Yuri Rzhanov, a research professor, has a Ph.D. in Physics and Mathematics from the Russian Academy of Sciences. He completed his thesis on nonlinear phenomena in solid state semiconductors in 1983. Since joining the center in 2000, he has worked on a number of signal processing problems, including construction of large-scale mosaics from underwater imagery, automatic segmentation of acoustic backscatter mosaics, accurate measurements of underwater objects from stereo imagery.

His research interests include development of algorithms and their implementation in software for 3D reconstruction of underwater scenes, automatic detection and abundance estimation of various marine species from imagery acquired from ROVs, AUVs, towed and handheld cameras.

Igor Kozlov received his B.S in Information Systems and Technologies from Saint Petersburg State Polytechnic University. He is currently pursuing a Master’s degree in Computer Science. His research interests include image processing and underwater image reconstruction.